W3C home > Mailing lists > Public > www-xkms@w3.org > July 2004

Re: Enumerations in schema

From: Berin Lautenbach <berin@wingsofhermes.org>
Date: Tue, 06 Jul 2004 21:01:50 +1000
Message-ID: <40EA869E.1030002@wingsofhermes.org>
To: tommy lindberg <lindberg_tommy@hotmail.com>
Cc: www-xkms@w3.org

Tommy,

My guess is that you are using Xerces-J?  (I'm using Xerces-C)

Strictly speaking, I don't think the behaviour below can be relied on in 
all parsers.  As I understand it, Xerces-C behaviour is perfectly 
acceptable for a schema validating parser.  So if we leave it as is, we 
are going to have cases where the messages cannot be validated because 
the parser is being very strict.

Cheers,
	Berin


tommy lindberg wrote:

> Berin, Rich
> 
>> I guess everyone tested using 'xkms' as the NS prefix.
> 
> 
> I didn't. :)
> 
> It seems to work ok for me; whatever prefix is bound to the XKMS 
> namespace must be used for the attributes and elements of type QName.
> 
> E.g. if I bind 'km' to the XKMS namespace and used 'xkms' as the prefix 
> for a KeyUsage value, this is what Xerces emits:
> 
> UndeclaredPrefix: Cannot resolve 'xkms:Signature' as a QName: the prefix 
> 'xkms' is not declared.
> The value 'xkms:Signature' of element 'km:KeyUsage' is not valid.
> 
> Regards,
> Tommy
> 
> 
>> From: Rich Salz <rsalz@datapower.com>
>> To: Berin Lautenbach <berin@wingsofhermes.org>
>> CC: "www-xkms@w3.org" <www-xkms@w3.org>
>> Subject: Re: Enumerations in schema
>> Date: Mon, 5 Jul 2004 12:29:27 -0400 (EDT)
>>
>>
>> >     <simpleType name="KeyUsageType">
>> >         <restriction base="QName">
>> >             <enumeration value="xkms:Encryption"/>
>> >             <enumeration value="xkms:Signature"/>
>> >             <enumeration value="xkms:Exchange"/>
>> >         </restriction>
>> >     </simpleType>
>> >
>> > I'm not a huge expert in XMLSchema, but my understanding is that
>> > enumeration values are literal.
>>
>> You are correct.
>>
>> Bummer. :)
>>
>> I guess everyone tested using 'xkms' as the NS prefix.
>>
>> I *hate* qname's as attribute values.  It is very late to change them to
>> URI's, although I know what WS-Security did it at last-call stage. WE
>> could then use Gudge's "open enumeration" technique to list the URI's 
>> that
>> are pre-defined by the standard.
>>
>>
>> *Something* has to be fixsed, either the schema enumeration values
>> removed, or QNAME->URI.  I vote for the second choice.
>>
>>     /r$
>>
>> -- 
>> Rich Salz                  Chief Security Architect
>> DataPower Technology       http://www.datapower.com
>> XS40 XML Security Gateway  http://www.datapower.com/products/xs40.html
>> XML Security Overview      
>> http://www.datapower.com/xmldev/xmlsecurity.html
>>
> 
> _________________________________________________________________
> FREE pop-up blocking with the new MSN Toolbar  get it now! 
> http://toolbar.msn.click-url.com/go/onm00200415ave/direct/01/
> 
> 
> 
Received on Tuesday, 6 July 2004 07:01:59 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 27 October 2009 08:39:22 GMT